I propose to engage in several activities, all designed to illuminate the etiology of neoplastic disease. I will supervise case-control studies in test of specific etiologic hypotheses suited to that method. The first will be of pancreas cancer in relation to diet and atopic disease, and of aggressive (metastatic or invasive through the myometrium) endometrial cancer in relation to exogenous estrogen use. I will continue to develop a registry of twins with cancer and perform co-twin studies of breast cancer, colon and other GI cancer, malignant melanoma, and other neoplasms. I will perform follow up on unaffected twins in relation to their stated exposures, and utilize population-based twin registries in California and Sweden for various adjunct studies. I will begin a cohort study in high risk individuals designed to elevate the determinants and implications of acquired benign pigmented lesions. I will continue to use cancer and registry information for descriptive studies, including estimation of the observed and expected rates of conjugal site-specific cancer, of the incidence among those residing near geographic foci of contamination with toxic wastes, and of disease-specific concordance in twins and the implications for interpretation of familiality and heritability. Finally, I will design a program to facilitate laboratory investigations making use of material from twins in the registry; such a program could provide a unique integration of high quality epidemiologic study design and "state of the art" lab techniques.